1 $Id: CHANGES,v 1.219 2003/03/06 14:35:48 stelian Exp $
3 Changes between versions 0.4b33 and 0.4b34 (released ?????????????????)
4 =======================================================================
6 1. Fixed the INSTALL file to reflect the actual install paths.
7 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
10 2. Fixed the configure script to only check for headers presence
11 instead of trying to compile them. This should fix issues
12 with old build environments. Thanks to Kari Lempiainen
13 <kari@funky.pp.fi> for reporting the bug.
15 3. Fixed restore to correctly ignore sockets when comparing
16 a dump (as socket cannot be properly restored anyway). Thanks
17 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
19 4. Fixed restore to correctly access the archive file (-A argument)
20 even when using a remote tape. Thanks to Fabrice Bellet
21 <fabrice@bellet.info> for reporting the bug.
23 5. Fixed (again) handling of long (largefile) seeks in rmt.
24 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
27 6. Fixed restore corner case when dealing with large block sizes
28 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
29 <fabrice@bellet.info> for reporting the bug.
31 7. Fixed a bug preventing dump to access a filesystem having
32 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
33 for reporting the bug.
35 8. Made dump store dump dates using explicit timezones, fixing a
36 problem with incremental dumps if the timezone is modified
37 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
38 the bug report and the patch.
40 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
41 =======================================================================
43 1. Added a note in the restore man page clarifying the question
44 'set the permissions on the current directory ?' asked by
45 restore at the end of treatment in -i and -x modes.
47 2. Fixed the endianess issues when reading compressed tapes.
48 Thanks to Dark Force <daq4th@users.sourceforge.net> for
49 reporting this bug and providing test cases.
51 3. Fixed the "ACL won't be dumped" warning message (which showed
52 an extra, unrelated error message). Thanks to Dragan Krnic
53 <dkrnic@lycos.com> for reporting this bug.
55 4. Made dump look first into /etc/mtab, then into /etc/fstab
56 when searching for filesystem paths. Also fixed some problems
57 caused by binding mounts or multiple block device mounts.
58 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
59 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
60 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
61 and #160305, Sourceforge bugs #618699 and #679832).
63 5. Made dump's -I option accept the value '0' meaning all the
64 read errors will be ignored. This can be useful when running
65 dump from unattended sessions (like cron jobs). Thanks to
66 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
68 6. Fixed the output of dump to indicate 'blocks' instead of
69 'tape blocks' in the various messages (blocks are always
70 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
71 made it clearly print the current blocksize at the start of
72 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
75 7. Made rmt understand long (largefiles) seeks.
77 8. Fixed build with very old versions of libext2fs, where
78 EXT2_FT_* constants were undefined.
80 9. Made dump accept the dumpdates path on the command line
81 (-D file option) instead of using only the hardcoded one.
82 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
85 10. Enabled rmt, LFS, readline, QFA options by default in
86 ./configure. Updated the configure process (new versions
87 of config.guess, config.sub etc).
89 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
90 =======================================================================
92 1. Changed dump to use fcntl(F_SETLK) style locking instead
93 of flock() when locking the dumpdates file. With the old
94 locking scheme, a local user having read rights on the
95 dumpdates file could be able to do a Denial of Service attack
96 on dump. In order to lock the dumpdates file with the new
97 scheme, the user would need to have write access on the file.
98 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
99 reporting the bug (originally a bugtraq post).
101 2. Fixed interactive 'ls' which caused spurious errors warnings
102 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
103 <jorgen_ostling@users.sourceforge.net> for reporting this
106 3. Fixed dump's estimate when dealing with sparse inodes.
108 4. Modified dump to allow setting a blocksize bigger than 32kB
109 (raised the limit to 1024kB), because newer hardware needs
110 this for proper operation. Thanks to Dirk Traenapp
111 <dtraenapp@users.sourceforge.net> for reporting this.
113 5. Fixed a bug causing Dump to stop and report an error if an
114 inode number in the exclude file was followed by some amount
115 of whitespace. Thanks to Jeffrey Sofferin
116 <sofferin@users.sourceforge.net> for reporting this bug.
118 6. Fixed a bug which caused restore, in some particular cases,
119 to ask some 'scary' questions and leave a bunch of RSTTMP
120 directories behind when restoring incremental tapes. Thanks
121 to Philippe Troin <phil@fifi.org> for reporting this bug and
122 providing the test cases.
124 7. Changed the wording when inodes are excluded from dump:
125 replaced 'Added inode 7 to exclude list' with
126 'Excluding inode 7 (resize inode) from dump', as suggested
127 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
129 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
130 ===================================================================
132 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
133 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
134 <eros@lamel.bo.cnr.it> for reporting the bug.
136 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
137 ===================================================================
139 1. Made dump print out the ext2 logical block number in case of a read
140 error, which can be given as an argument to debugfs' ncheck command,
141 in order to find out the affected inode. Added note about this
142 usage in dump's man page.
144 2. Fixed a problem in restore when reading tapes written on big
145 endian machines with very old versions of dump. The patch was
146 contributed by George Helffrich <george@geology.bristol.ac.uk>.
148 3. Fixed the tape length calculation when using large tapes
149 and compression. Thanks to Georg Lippold
150 <g_lippold@sourceforge.net> for reporting the bug.
152 4. Added a new set of examples in dump_on_cd_2 directory, based
153 on dump_on_cd examples but somewhat enhanced, supporting DVD
154 media, and localized in english and german. Thanks to
155 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
157 5. Made dump save 32 bit UID/GID. Internally, this was achieved
158 by switching from the old BSD inode format to the new BSD
159 inode format, which means that the tape format was changed.
160 However, since all restore versions out there should
161 transparently support both inode formats, the change should
162 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
165 6. Fixed a lot of warnings in the code shown when compiling
166 with 'gcc -W'. Thanks to Matthias Andree
167 <matthias.andree@stud.uni-dortmund.de> for reporting this.
169 7. Fixed a small markup bug in the dump man page. Thanks to
170 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
173 8. Rewrote entirely the man pages using the tmac.an macro
174 package (Linux man page format) instead of the original BSD
175 format. They should be now cleaner and easier to modify.
177 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
178 ==================================================================
180 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
181 non Linux clients were misinterpreted. The description of the problem
182 (incompatible numbering in Linux mtio opcodes) is documented at
183 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
184 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
185 providing an excellent, cross-platform replacement for rmt in his
188 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
189 which caused restore to fail when the inode maps were not entirely
190 contained on the first volume (when using really small volumes or
191 when dumping a huge number of inodes).
193 3. Fixed a problem in dump, where files in subdirectories of directories
194 that have the nodump flag set (or being excluded with -e/-E) were
195 sometimes incorrectly dumped (instead of being excluded). The original
196 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
197 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
198 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
199 help testing the patch.
201 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
202 which makes one able to pipe the output of dump, by the net, to
203 a remote CD-burner server.
205 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
206 and use GNU's symbolic syntax over rmt instead of numerical values
207 to assure multiple platform compatibility.
209 6. Documented the -d option in restore.
211 7. Added a -v (verbose) mode to dump. For now it just prints the number
212 of the inode being dumped, but this could evolve in future versions
213 to include interesting debugging output.
215 8. Added a -o flag to restore, which automatically restores the current
216 directory permissions in -i or -x mode, without asking the operator
217 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
218 and Peter Samuel <peters@e-smith.com>.
220 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
221 ====================================================================
223 1. Fixed a bug in the QFA feature which could have caused
224 a bad tape positionning on restore (causing extra delays in
225 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
227 2. Added a small note in the dump man page specifying that
228 there should be no white space between the option letter and
229 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
231 3. Made restore work with older versions of the readline library,
232 by checking in configure for several extended symbols. Restore
233 can now be compiled with a readline as old as the 2.0 release
234 (though it may be a good idea to upgrade it to a more recent
235 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
236 reporting the build failures.
238 4. Fixed a performance problem with the QFA file creation in
239 dump, which made unnecessary seeks on the tape slowing down
240 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
241 reporting this issue.
243 5. Removed the inclusion of some kernel headers in the dump
244 source, which prevented the compile in some kernel/glibc
245 headers/architecture combination. Thanks to Bdale Garbee
246 <bdale@gag.com> for reporting the bug.
248 6. Added the appropriate error message when dump fails to
249 open the output file for writing. Thanks to Amith Varghese
250 <amithv@yahoo.com> for reporting this bug.
252 7. Made restore able to understand large Solaris ufsdump tapes
253 (containing inodes bigger than 4194304). Sun have introduced
254 an "extension" to the dump tape format when dealing with
255 those inodes, which was uncorrectly handled by Linux restore.
256 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
257 providing a test case.
259 8. Added the -m parameter to dump which optimises the output for
260 inodes having been changed but not modified since the last dump
261 ('changed' and 'modified' have the meaning defined in stat(2)).
262 For those inodes, dump will save only the metadata, instead of
263 saving the entire inode contents. Inodes which are either
264 directories or have been modified since the last dump are saved
265 in a regular way. Uses of this flag must be consistent, meaning
266 that either every dump in an incremental dump set have the flag,
269 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
270 =======================================================================
272 1. Fixed behaviour of dump when exceeding resource limits
275 2. Added the -L flag to restore to allow the user to specify a
276 maximal allowed number of miscompares when using restore
277 with the -C option to check the backup.
279 3. Detailed the manual entry for the -N option of restore.
281 4. Added the -a flag to restore to make able doing unattended
282 restores in -i or -x mode (automatically walks through the
283 multiple dump volumes).
285 5. Extended the QFA mode to work with local files and/or
286 remote tapes and files. This way, restore can know in advance
287 the tape number and the offset for the inodes to extract and
288 can minimize the extraction time by seeking directly to the
291 6. Added the -A <archive> option to both dump and restore,
292 which makes dump to archive a dump table-of-contents in
293 the specified file to be used by restore to determine
294 whether a file is in the dump file that is being restored.
295 (the archive file syntax is also compatible with the
296 Solaris ufsdump generated one).
298 7. Small fix in restore making it able to read some (broken ?)
299 Solaris ufsdump tapes.
301 8. Fixed dump to correctly recognise the root filesystem when using
302 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
303 for reporting this bug.
305 9. Added the -P <file> option to restore to create a
306 Quick File Access file from an already made dump. Patch
307 contributed by Uwe Gohlke <uwe@ugsoft.de>.
309 10. Made restore compile and run on Solaris, making it a
310 possible replacement for the standard ufsrestore. Port was
311 contributed by Uwe Gohlke <uwe@ugsoft.de>.
313 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
314 =====================================================================
316 1. Added a set of backup scripts from Eugenio Diaz
317 <getnito@yahoo.com> in the examples section. It features
318 automatic (cron based) full and incremental dumping of
319 several filesystems on a separate filesystem.
321 2. Fixed a off-by-one miscalculation which disabled dumping
322 a one letter subdirectory. Thanks to Chris Siebenmann
323 <cks@utcc.utoronto.ca> for reporting the bug.
325 3. Fixed several restore bugs occuring when trying to
326 restore some missing files on the tape. Thanks to Chris
327 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
329 4. Fixed --with-ldopts configure argument passing, installing from
330 a separate object directory, makefile cleanups contributed
331 by <splite@purdue.edu>.
333 5. Fix a bug which could caused, in some conditions, the highest
334 number inode of a filesystem, to not be dumped. Many thanks
335 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
338 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
339 =======================================================================
341 1. Added a mini howto from Patrick Walsh in the examples
344 2. Minor man pages syntax corrections. Thanks to
345 Chris Danis <screechco@home.com> for reporting the bugs.
347 3. Added a script from David B. Peterson <dave@toppledwagon.com>
348 to the examples section. It features dumping several
349 filesystems to a remote tape drive upon ssh.
351 4. Added a patch provided by Richard Jones <rich@annexia.org>
352 which allows BRADEMAX (number of read errors tolerated by
353 dump) to be adjusted using the -I option.
355 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
356 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
359 6. Add the -l option to restore to specify if, when doing a
360 remote restore, the file used to access the backup is a
361 regular file (the defaults being a tape drive). Restore needs
362 to know this information when reading a remote compressed
363 dump. Previously, this information was autodetected, but
364 the autodetection code fails (with ioctl: Inappropriate ioctl
365 for device) when using a non Linux remote box. Thanks to
366 many users and especially to Eros Albertazzi
367 <eros@lamel.bo.cnr.it> for reporting this.
369 7. Found a workaroung for the dump deadlock problem (3 childs
370 stuck in pause(), father in read()). The workaround seems
371 to work for me and several beta-testers. If it doesn't work
372 for you, please report back.
374 8. Updated the RPM spec file (BuildPrereq, URL etc).
376 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
377 ========================================================================
379 1. Fixed the permissions of a newly created QFA file by dump.
381 2. Cleaned up the source of dump (the external variables
382 definition was a complete mess, making possible to have
385 3. Fixed restore to use the full tape volume path when doing
386 a compare (since it changes the working directory to the
387 filesystem being compared in the process).
389 4. Added the -q option to dump which makes dump abort
390 whenever it needs operator attention. It should be
391 easier to use dump in scripts with this option.
393 5. Detect the use of incompatible options to dump and
394 refuse them (like -a and -B options together).
396 6. Added bzip2 compression to dump/restore (use option -j level
397 to select it). Note that you will need a restore version
398 >= 0.4b24 in order to restore a bzip2 compressed dump.
399 The same warning as for the zlib compression applies:
400 the tape format of a bzip2 dump is not compatible with the
401 original BSD tape format.
403 7. Fixed a overflow problem in dump corrupting the dump when
404 very large files were encountered. Thanks to Vladimir Nadvornik
407 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
408 kernel buffer/page cache before starting the dump, helping
409 a bit those who use dump on mounted filesystems. Thanks to
410 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
413 9. Updated the RPM spec file following the RedHat changes
414 (dynamically linked binaries now in /usr/sbin etc).
416 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
417 which enables restore to recognise multi volume compressed dumps
418 done on CD. Included his scripts for dump (which pipe the dump
419 output directly into cdrecord) and restore. There is now
420 possible to to backups to CD on the fly!
422 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
423 ===================================================================
425 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
426 Marcus Meissner (Caldera International Security Dept.).
428 2. Implement the Sun rmt extended protocol. Patch submitted
429 by Ian Gordon <iangordon@users.sourceforge.net>.
431 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
432 the linux kernel header. This ensures that dump always has the
433 latest version of this file. Patch submitted by
434 Andreas Dilger <adilger@turbolinux.com>.
436 4. Report any filesystem present in either /etc/fstab with a
437 positive passno or /etc/dumpdates in dump -w output.
438 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
440 5. Fixed the looping problem in dump introduced in the
443 6. Changed the -B option of dump to limit the size of
444 _compressed_ output per volume if compression is on.
445 Patch contributed by Helmut Jarausch
446 <jarausch@igpm.rwth-aachen.de>. Note however that, since
447 it is impossible to predict the size of the compressed
448 data before doing the compression, the -B limit is a bit
451 7. Fixed a bug in reading the operator typed file/tape path for
452 the next volume in restore.
454 8. Implemented a "-F script" option to restore which permits the
455 user to specify a script which will be launched at the
456 beginning of each tape, useful for automatic programming of
457 tape changers for example. See the restore man page for the
458 script parameters and return codes.
460 9. Small fix for the QFA routines provided by Uwe Gohlke
461 <uwe@ugsoft.de>, and some recommendations for QFA uses in
464 10. Fixed the multivolume restoring where making a mistake
465 to the 'Mount next tape' prompt caused several blocks to
468 11. Enhanced the -e option of dump to take as a parameter a
469 comma separated list of inode numbers.
471 12. Added the -E option to dump which specify a file containing
472 inode numbers to exclude from the dump.
474 13. Fixed the compressed multi-volume dump + restore.
476 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
477 ==================================================================
479 1. Made dump report the number of blocks dumped per volume.
480 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
482 2. Fix a bug in dump where waiting too much at the 'change volumes'
483 question caused the volume to be rewritten. Thanks to
484 Graham Reed <greed@users.sourceforge.net> for reporting the
485 bug and providing a patch.
487 3. Added a compression option to dump, contributed by Jerry
488 Peters <gapeters@worldnet.att.net>.
490 WARNING: the compression works only on files/pipes and
491 drives supporting variable block size.
493 WARNING2: tapes created using the compression feature are
494 incompatible with the standard dump tape format, and a
495 version of restore >= 0.4b22 will be needed for extraction.
497 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
498 architectures. Thanks to Paul Slootman <paul@debian.org> for
499 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
502 5. Many cleanups (CPP defines, const char warnings, check of
503 ext2fs COMPAT flags, time_t cleanups, added libext2 version
504 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
506 6. Made --prefix option work in configure. All the install path
507 are now based on the configure parameters.
509 7. Added the Quick File Access mode in dump/restore, contributed
510 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
511 a file tape position for each inode, and this file is used by
512 restore (if called with parameter Q and the filename)
513 to directly position the tape at the file restore is currently
514 working on. This saves hours when restoring single files from
515 large backups, saves the tapes and the drive's head. Use
516 --enable-qfa option of configure to compile in the QFA support.
518 8. Added the possibility to dump several files and directories
519 in a single invocation of dump. Thanks to Uwe Gohlke
520 <uwe@ugsoft.de> for implementing this option.
522 9. Fixed the dumping and restoring of files with holes
523 (files ending with holes didn't get dumped or restored
526 10. Fixed a socket descriptor leak in dump, which leaved opened
527 3 file descriptors per dump process (and there is one dump
530 11. Fixed dump large file system support, by explicit use of
531 open64/lseek64/etc functions (explicit use needed because
532 e2fsprogs libraries don't behave well when compiled with
533 FILE_OFFSET_BITS=64).
535 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
536 ======================================================================
538 1. Fixed some bugs in the dump -w|-W logic introduced by
539 the previous version. Thanks to Andreas Dilger
540 <adilger@turbolinux.com> for his help on this one.
542 2. Fixed again a compilation problem when using old e2fs
543 headers (filesystem label related). Thanks to many users
544 who reported this stupid error.
546 3. Fixed a build problem on old lib5 based systems dealing with
547 _PATH_MNTTAB being undefined. Thanks to John Adams
548 <johna@onevista.com> for reporting the bug.
550 4. Improved the error detection in restore when creating
551 the temporary files in TMPDIR. Now restore will corectly
552 report a 'no space left on device' error instead of
553 strange errors which could imply an archive corruption.
554 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
555 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
557 5. Added the throughput information in dump's progression
558 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
561 6. Use libext2fs's inode scanning routines, which are particularly
562 robust when dealing with errors (especially when having some
563 bad blocks on the filesystem). Many thanks to Colin
564 <colin@horizon.com> for reporting the bug, investigating
565 the issues, finding a workaround, writing the patch and
566 fully testing it... (of course, if this patch will break
567 anything, he is to blame for :-)).
569 7. Made dump and restore LFS aware. Dump can dump a filesystem
570 containing large files, generate a large file on output and
571 restore can restore them. This feature is not enabled by
572 default, use --enable-largefile option of configure to enable
573 it (you need to have a LFS aware glibc though). Thanks to
574 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
575 and to Theodore T'so <tytso@valinux.com> for his always
578 8. Made dump ask upon a tape write error if it should rewrite
579 the current volume (assume this is a bad tape) or if it should
580 assume an end-of-tape condition (useful for tape drives which
581 behaves badly at the end of the tape). Thanks to Andreas
582 Dilger <adilger@turbolinux.com> for the suggestion.
584 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
585 =======================================================================
587 1. Fixed a small compilation problem due to a change
588 in the definintion of the struct sigaction in
589 glibc 2.0 / libc5. Thanks to Gunther Schlegel
590 <schlegel@riege.de> for reporting the bug and to
591 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
594 2. Modified the label and uuid analysis in order to be
595 self-contained (independant of kernel/libc headers). This
596 should fix the compile with older kernel/libc headers and
597 will preserve the functionality. Thanks to Bernhard Erdmann
598 <bernhard.erdmann@gmx.de> for reporting the bug.
600 3. The 'exclude inode' option, if applied to a directory
601 inode, excludes now correctly all the leaves of this
602 directory. Thanks to John R. Dennison
603 <gerdesas@users.sourceforge.net> for reporting the bug.
605 4. Fixed the '-e' option to disable the possibility
606 to exclude the root inode (which causes the dumps to
607 be unreadable by restore). Prevented array overflow
608 when multiple -e options are used.
610 5. Fixed dump to correctly interpret a filesystem argument
611 which is a mountpoint and it is not an absolute path
612 (as specified in the fstab). Thanks to Bernhard R. Erdmann
613 <be@berdmann.de> for reporting the bug.
615 6. Made dump able to backup files larger than 2 GB. Note that
616 dump still doesn't cope with files larger than 4 GB.
618 7. Restore the real uid and gid before invoking an external
619 RSH program (big hole when dump or restore is suid root!).
621 8. Get the values for the default tape device and the location
622 of fstab file from the system headers. Thanks to
623 Andreas Dilger <adilger@turbolinux.com> for the patch.
625 9. Made dump -w|-W report all recognized filesystems
626 present in either /etc/fstab or /etc/dumpdates, and present
627 the list in the same order as in fstab file. Thanks
628 to Andreas Dilger <adilger@turbolinux.com> for the patch.
630 10. Made dump's -a (automatic end of tape detection) the
631 default. Specifying one of -B, -c, -d or -s options will
632 override the EOM detection. Thanks to Andreas Dilger
633 <adilger@turbolinux.com> for the patch.
635 11. Save the ext2 filesystem label into the dump volume label.
636 Specifying a volume label on the command line (-L option)
637 will override this feature. Thanks to Andreas Dilger
638 <adilger@turbolinux.com> for the patch.
640 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
641 =====================================================================
643 1. Fixed the signal handling in dump (which I broke in 0.4b17)
644 which was causing several strange dump failures (dump
645 hanged or segmentation faults etc).
647 2. Specified the default blocksize in the dump man page.
649 3. Changed two info messages of restore to be written on stdout
650 instead of stderr in order to leave stderr only for errors
651 or warnings. Thanks to Stephen Atwell
652 <satwell@urbana.css.mot.com> for the suggestion.
654 4. Corrected an off by one calculation which prevented
655 dumping a directory having a 1 character name length.
656 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
657 for reporting the bug.
659 5. Reinforce dump in order to allow only level 0 and no
660 -u option when dumping a subdirectory, and document
661 this in the man page. Thanks to Bernhard Erdmann
662 <bernhard.erdmann@gmx.de> for reporting the bug.
664 6. Fixed a small harmless bug in restore which caused
665 in some conditions hard links to be restored several
666 times (therefore generation some warning messages).
667 Thanks to Kenneth Porter <shiva@well.com> for
670 7. Updated the RPM spec file to the latest RedHat version,
671 providing FHS packaging and other cosmetic changes.
672 You will need RPM version >= 3.0.5 in order to build the RPM.
674 8. Updated the configure script to check for libtermcap
675 before checking for libreadline, because we need this
676 library in order to compile the readline support.
678 9. Made dump understand the LABEL= and UUID= notation
679 both in /etc/fstab and on the command line. Note that
680 you will need the /proc filesystem in order to use
681 these notations. Thanks to Erik Troan <ewt@redhat.com>
682 for providing the patch.
684 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
685 ===================================================================
687 1. Fixed a potential buffer overflow in restore. Thanks
688 to Stan Bubrouski <satan@fastdial.net> for reporting
691 2. Fixed a readline-related bug which prevented
692 'cat DUMPFILE | restore -i -f -' from working. Thanks
693 to Charles Karney <karney@users.sourceforge.net>
696 3. Changed a few "panic" into "exit", causing restore to
697 be more stable against some attacks (like the last one
698 reported on Bugtraq, although the last version of restore
699 was not vulnerable - just dumped core). Thanks to
700 Andreas Hasenack <andreas@conectiva.com.br> for reporting
703 4. Removed the suid-root bit on dump and restore in the
704 default build (and generated RPMs). It should be safer
705 now. Users who need the suid-root capabilities in order
706 to make network backups should read first the man page
707 of dump and enable the suid bit manually.
709 5. Added -ltermcap to the compile parameters for restore
710 when using readline support, in order to make the compile
711 process work on some systems (SuSE ?). Thanks to
712 Patrik Schindler <poc@pocnet.net> for reporting the bug.
714 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
715 ====================================================================
717 1. The -F script is called now *only* at the end of a tape,
718 not at the end of the dump. Thanks to Christian Haul
719 <haul@informatik.tu-darmstadt.de> for the bug report.
721 Normally, the device name and the current volume number
722 are passed on the command line of the script. However,
723 if you want the old style script (with no arguments
724 passed on the command line), you can enable it in
725 configure with the --enable-oldstylefscript.
727 2. Use posix signal handling to preserve dump functionality
728 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
731 3. Made the exit code of restore in case of a 'C'ompare
732 command reflect the result of the compare. An exit status
733 of 0 means the dump archive is correct, 1 means tape errors,
734 2 means that some files were modified. Thanks to Kenneth Porter
735 <shiva@well.com> for the suggestion.
737 4. Made (finally) quotes work in interactive restore.
739 5. Small fixes in order to allow dump to compile with a
740 really old e2fsprogs version. Thanks to Ian Zimmerman
741 <itz@speakeasy.org> for the bug report.
743 6. Add GNU readline capabilities to interactive restore.
744 Use configure's --enable-readline flag to enable this feature.
745 Thanks to Patrik Schindler <poc@pocnet.net> for the
748 7. Do the compare on the fly in restore 'C' mode (this will
749 allow not to exhaust the available /tmp space when
750 ccmparing large files). Thanks to Kenneth Porter
751 <shiva@well.com> for the suggestion.
753 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
754 ====================================================================
756 1. Fixed some several duplicate 'DUMP: DUMP:' in the
759 2. Corrected the estimation of blocks to dump. Note that
760 this is still wrong for large directory inodes, due
761 to the size of a BSD directory entry on the tape
762 (estimation is based on the size of the ext2 directory,
763 which is smaller because it doesn't contain the null
764 character at the end).
766 3. Made dump report the total number of MB written to
767 tape at the end of run. Thanks to W. Reilly Cooley
768 <wcooley@nakedape.cc> for the patch.
770 4. Added the -X option to restore giving the possibility
771 to read the names of the files to be extracted/listed
772 from a text file (in addition of the command line).
773 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
776 5. Added the device name and the current volume number
777 as arguments to the end of tape script (dump -F option).
779 6. Made the multi-volume dump work again (after having
780 broken it in 0.4b15).
782 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
783 ===================================================================
785 1. Added a prompt command in interactive restore mode. Thanks
786 to Andreas Dilger <adilger@home.com> for the patch.
788 2. Fixed a buffer overflow problem in dump (caused by
789 not checking the size of the filesystem parameter).
790 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
791 reporting this on Bugtraq (and to several dump users
792 who forwarded me his mail).
794 3. Added the '-F script' option to dump in order to
795 launch a script at the end of each tape (to be used
796 with a tape changer, or to notify the sysadmin by
799 4. Fixed a bug in restore compare code caused by the changes
802 5. Fixed the treatment of options using the old BSD syntax
803 in both dump and restore.
805 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
806 =======================================================================
808 1. Fixed a bug in dump which may have caused invalid deleted
809 directories to be dumped out if they were deleted after the
810 mapping pass. This could occure on active filesystem and lead
811 to heap corruption (causing dump malfunction in many possible ways).
812 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
813 bug and submitting the patch.
815 2. Fixed the handling of the filesystem parameter in dump. A
816 '/mnt/disk' parameter caused the disk contents to be dumped,
817 but a '/mnt/disk/' parameter caused the mountpoint directory
818 to be dumped (generally an empty directory).
820 3. Improved the output of dump in order to tell which directory
821 it is currently dumping (when dumping a subtree).
823 4. Added the '-e' exclude inode option to dump. Thanks to
824 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
826 5. Added a REPORTING-BUGS file in order to provide a guide
827 on how to correctly report a bug in dump/restore.
829 6. Fixed a restore problem when restoring a hard link to an inode
830 having the immutable or append-only attribute set. Thanks to
831 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
833 7. Fixed a compatibility problem between dumps made on little
834 endian machines (the format was correct) and big endian
835 machines (the format was incorrect). This fix break the
836 compatibility with the older dumps made on big endian
837 machines (sparc, m86k, ppc etc). For the first time in
838 linux dump's history (I believe), the dumps made by *BSD,
839 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
840 so interchangeable. Thanks to Rob Cermak
841 <cermak@ahab.rutgers.edu> for submitting the bug and
842 helping me test the fix.
844 8. Fixed the way dump reports the remaining percent/time, if
845 the number of blocks actually dumped exceeds the estimated
846 number of blocks. Thanks to Jean-Paul van der Jagt
847 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
849 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
850 ======================================================================
852 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
853 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
855 2. Removed the requirement to build the RPM as root from the
856 spec file. Thanks to Christian Weisgerber
857 <naddy@mips.rhein-neckar.de> for submitting this.
859 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
860 causing dump to block when dumping really huge filesystems.
861 Many thanks to Patrik Schindler <poc@pocnet.net> for
862 helping me find this bug.
864 4. Fixed the treatment for an interrupt signal when dump access
865 the remote tape through RSH. Thanks to Christian Weisgerber
866 <naddy@mips.rhein-neckar.de> for providing the patch.
868 5. Fixed a bug which was causing dump/restore to display
869 garbage characters instead of the remote host name.
871 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
872 =====================================================================
874 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
875 <kukuk@suse.de> for submitting the patch.
877 2. Fix for the exit code when using the size estimate option of
878 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
879 submitting the patch.
881 3. Handle EINTR in atomical reads/writes in dump, which was causing
882 dump to fail on some systems. Thanks to Eric Jergensen
883 <eric@dvns.com> for reporting the bug and submitting the patch.
885 4. Handle more than 16 characters for the device names in dumpdates.
886 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
887 tracking down the problem and proposing the solution.
889 5. Fixed a bug in dump which prevented the creation of the
890 dumpdates file when doing a 0-level dump without already
891 having a dumpdates file. Thanks to Patrik Schindler
892 <poc@pocnet.net> for reporting the bug.
894 6. Changed the way dump 'S' flag reports the size estimate
895 from number of blocks into bytes (making it compatible
896 with the Solaris version, and simplifying things for
897 amanda users). Thanks to Jason L Tibbitts III
898 <tibbs@math.uh.edu> for reporting the bug.
900 7. Fixed a compatibility problem in linux/alpha dump tape format.
901 Now the linux/alpha dump are (again) compatible with the
902 other dump formats. But this breaks compatibility with
903 older dumps made on alpha. Thanks to Mike Tibor
904 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
906 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
907 ======================================================================
909 1. Added a '--enable-kerberos' to configure.
911 2. Added a 'S' option to dump which determines the amount of space
912 that is needed to perform the dump without actually doing it, similar
913 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
914 <cermak@ahab.rutgers.edu>.
916 3. Added a 'M' multi-volume option to dump and restore which enables
917 dumping to multiple files (useful when dumping to an ext2
918 partition to make several dump files in order to bypass the 2GB
919 file size limitation). The 'f' argument is treated as a prefix and
920 the output files will be named <prefix>001, <prefix>002 etc. With
921 the 'M' flag, restore automatically selects the right file without
922 asking to enter a new tape each time.
924 4. Fixed a memory leak which was causing dump to grow very big
925 (270MB when dumping a 10GB filesystem...). Thanks to Jason
926 Fearon <jasonf@netrider.org.au> for reporting the bug.
928 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
929 ======================================================================
931 1. Make configure test if the system glob routines support
932 extended syntax (ALTDIRFUNC). If not, use the internal glob
933 routines instead of system ones. Thanks to Bernhard Sadlowski
934 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
935 and helping me resolve this and other minor libc5 compiling
938 2. Fix a problem when dumping a ext2fs with the 'filetype'
939 feature enabled. Thanks to Patrick J. LoPresti
940 <patl@cag.lcs.mit.edu> for reporting the bug and to
941 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
943 3. Made the nodump flag work on directories. A directory which
944 has the nodump flag gets never dumped, regardless of its
947 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
948 which allows dump on an active ext3 filesystem. However, this
949 is a "quick and dirty" patch which enables backup of an ext3
950 filesystem through the ext2 compatibility (by ignoring the
951 NEEDS_RECOVERY bit). The journal file is not recognized and
952 it is dumped (it should not).
954 5. Test the superblock compatibility flags when dumping, in order
955 to be sure that we know how to deal with specific features.
957 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
958 ====================================================================
960 1. Use lchown instead of chown, fixing a possible security problem
961 when restoring symlinks (a malicious user could use this
962 to deliberately corrupt the ownership of important system files).
963 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
964 this and providing the patch.
966 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
967 ====================================================================
969 1. Put dump sources under CVS, added Id tags in all files so
970 one can use 'ident' on binary files.
972 2. Added the dump/restore version in the usage text so one can
973 easily verify the version he is using.
975 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
976 a va_start/va_end problem on linux-ppc (always call va_start
977 va_end in pairs each time we use a vararg function).
979 4. Added again the DT_* constants because old libc does not
980 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
981 for submitting the bug report.
983 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
984 this should enable dumping big (huge) filesystems.
986 6. Added the RSH environment variable in order to be able to
987 use a rsh replacement like ssh when doing remote backups (and
988 bypass the security limitations of rcmd). Now you can do remote
989 backups without being root (or making dump setuid root).
991 7. Modified again the way dumpdates works. For incremental dumps,
992 we need to read dumpdates even if we are not using 'u' option.
993 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
996 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
997 ===================================================================
999 1. Removed the 'k' flag from the restore 'about' text if kerberos
1000 was not compiled in.
1002 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1003 to them (fsetflags takes a char*, setflags an open fd!).
1005 3. (f)setflags is called only if the flags aren't empty. If the
1006 file is a special file, a warning is printed, because changing
1007 flags implies opening the device. Normally, a special file
1008 should not have any flag... (Debian bug #29775, patch provided
1009 by Abhijit Dasgupta <abhijit@ans.net>).
1011 4. Made possible to dump a file system not mentioned in /etc/fstab.
1012 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1014 5. Changed the default behaviour to not create dumpdates
1015 unless 'u' option is specified. Removed the old "debian-patch"
1016 which provided the same thing. (Debian bug #38136, #33818).
1018 6. Removed all those dump*announce, since they were getting old...
1020 7. Added warning messages if dumpdates does not exist and
1021 when an inferior level dump does not exist (except for a level 0
1024 8. Debugged the glob calls in interactive mode: restore used a
1025 dirent struct which was different from the /usr/include/dirent.h
1026 one (this used to work, is it a glibc2 change?), so none of the
1027 compat glob (which used /usr/include/dirent.h) or the system glob
1028 worked. Restore use now the system dirent (and the system
1029 DT_* constants), which are compatible with BSD ones.
1031 9. Added a configure flag (--with-dumpdatespath) to specify
1032 the location of dumpdates file. By default, it is
1035 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1036 included the current date/version in man pages.
1038 11. Corrected the estimation of remaining time when
1039 the operator doesn't change the tapes quickly enough. This
1040 was an old bug, I thought I corrected it, and discovered
1041 that in fact it was corrected in two different places, so
1042 the results canceled each other...
1044 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1045 ===================================================================
1047 1. Integrated multiple patches from RedHat, Debian and SuSE:
1049 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1051 - several fixes in the man pages.
1052 - update the default tape device to /dev/st0.
1053 - many updates for Linux Alpha (byte ordering, size_t etc).
1055 - use environment variable for TMPDIR (instead of /tmp).
1056 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1057 - workaround egcs bug (RedHat bugs #4281 and #2989).
1058 - wire $(OPT) throughout Makefile's.
1060 2. Upgrade the dump revision to 1, making possible to dump filesystems
1061 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1063 3. Fix some compile warnings, prototype all functions.
1065 4. Use glibc err/glob instead of internal compatibility
1066 routines (only if available).
1068 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1069 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1071 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1072 ======================================================================
1074 1. Integrated the changes from FreeBSD-3.1-RELEASE
1075 (mostly bug fixes, buffer overruns, dump has now an "automatic
1076 tape length calculation" flag, dump/restore can use kerberos now
1077 (this is NOT tested), use environment variables for TAPE and
1080 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1082 3. Corrected a bug in restore when using 'C' option with multi-volumes
1083 tapes (files splited accros two tapes give "size changed" errors
1086 4. Corrected the long standing bug when dumping multiple tapes.
1087 This works for me, needs further testing.
1089 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1090 ====================================================================
1092 1. Dump now runs correctly on kernels 2.1.x
1093 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1095 Changes between versions 0.4b2 and 0.4b3
1096 ========================================
1098 1. Use realpath() if available
1100 2. Report statistics
1102 Changes between versions 0.4b1 and 0.4b2
1103 ========================================
1105 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1108 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1110 3. Fixed configure.in to correctly handle --enable-rmt
1112 Changes between versions 0.3 and 0.4b1
1113 ======================================
1115 1. Integrated the changes from 4.4BSD-Lite2
1117 2. Integrated the patches from Debian and Red Hat
1119 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1121 4. Changed dump to use the Ext2fs library to get block addresses. This
1122 should solve the endianness problem on SparcLinux.
1124 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1125 distribution's one) to use autoconf
1127 6. Fixed a few minor bugs
1129 Changes between versions 0.2e and 0.2f
1130 ======================================
1132 1. Added the creation of named pipes (fifos) in restore.
1134 2. Added the -N flag in the restore manual page.
1136 3. Added the file kernel.patch which contains the llseek() optimization
1137 patch for 1.2.x kernels.
1139 4. Fixed a bug in the restoration of symbolic links: owner and group were
1142 5. Integrated some changes from FreeBSD 2.2.
1144 6. Added a call to ftruncate() after restoring each file to restore
1145 correctly files ending by a hole.
1147 Changes between versions 0.2d and 0.2e
1148 ======================================
1150 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1151 with owner = root (0) and group = root/wheel/whatever (0).
1153 Changes between versions 0.2c and 0.2d
1154 ======================================
1156 1. Dump is now able to backup 2GB+ filesystems.
1158 2. Dump and restore can now be linked as static binaries.
1160 Changes between versions 0.2b and 0.2c
1161 ======================================
1163 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1165 Changes between versions 0.2a and 0.2b
1166 ======================================
1168 1. Really fixed the bug that I should have corrected in 0.2a.
1170 2. Enabled optimization again.
1172 Changes between versions 0.2 and 0.2a
1173 =====================================
1175 1. Disabled the optimization during compilation.
1177 Changes between versions 0.1 and 0.2
1178 ====================================
1180 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1181 the fs_type field (actually, I modified the file fstab.c to make it
1182 use the mntent functions).
1184 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1185 made by dump should be readable by the BSD restore and backups made
1186 by the BSD dump should be readable by restore. Unfortunately, this
1187 means that the dump format has changed between version 0.1 and version
1190 3. Dump is now able to backup a subtree, it is no longer limited to whole
1191 filesystems like the BSD version.
1193 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1196 Changes between versions 0.0 and 0.1
1197 ====================================
1199 1. Now create links rdump and rrestore during the `make install' step.
1201 2. Linux port specific bugs added to the manual pages
1203 3. Incorrect estimation of the number of tapes blocks fixed when doing
1204 incremental backups.
1206 4. Better ls-like format in restore in interactive mode.